Estrogen receptor-? positive (ER?+) breast cancer (BC) represents 70-80% of newly diagnosed cases. While potentially responsive to anti-estrogens, progression of ER?+ BC to anti-estrogen refractory disease in the metastatic setting is a common occurrence. Endocrine resistance mechanisms may differ between subtypes of ER?+ BC and involve both ER?-independent and ER?-dependent forms that remain poorly understood. Prolactin (PRL) interacts with estrogens to stimulate growth of certain subtypes of ER?+ BC. At the same time, the PRL-Jak-Stat5 pathway promotes differentiation and inhibits invasive features of BC, and loss of Stat5a signaling in a subgroup of ER?+ BC is associated with anti-estrogen therapy failure. The involvement of PRL pathways in growth and progression of anti-estrogen refractory BC subtypes remains to be determined. There is a lack of preclinical human ER?+ BC models that recapitulate progression from localized mammary gland growth to distant metastasis. Xenografts of patient-derived ER?+ BC exhibit poor take rate in mice. We have discovered that murine PRL is a poor agonist and a potent antagonist for human PRL receptor (PRLr). Bovine PRL is also a poor agonist with antagonist activity for human PRLr. Laboratory human BC lines therefore have been selected for PRL-independent growth and may only represent subtype(s) of ER?+ BC. To address this problem, we generated hPRL knock-in mice in the immunodeficient Nod-Scid-IL2R? strain that express physiological levels of circulating hPRL. Remarkably, PRL-humanized mice display greatly increased take rate of patient-derived xenografts of ER?+ BC. We established a panel of serially transplantable ER?+ Luminal B lines, several of which spontaneously metastasize to lungs and liver. The distant metastases become anti-estrogen refractory despite continued expression of ER?, but show PRL-dependence. Our long-range goal is to determine mechanisms of anti-estrogen refractoriness of BC to improve clinical management. Aim 1 is focused on Luminobasal subtype of BC (ER?+/CK5+) and explores a distinct ER?-independent mechanism of anti-estrogen refractory BC recently recognized in pre-existing ER?+ laboratory cell lines. We hypothesize that in Luminobasal BC, loss of PRL-Stat5 signaling promotes loss of ER? and subsequent anti-estrogen resistance due to defective Stat5a-driven differentiation. Aims 2 and 3 are focused on the Luminal B BC subtype (ER?+/CK5-/Ki67high) and will use PRL-dependent patient-derived xenograft lines. Aim 2 will test the hypothesis that in Luminal B BC, unlike in Luminobasal BC, PRL facilitates growth and survival of metastases and that PRLr-pathway targeting cooperates with anti-estrogens to eliminate distant metastases. Aim 3 will explore novel strategies for detecting and monitoring latent residual disease.